What is data value?

It’s now widely understood that enterprise data is an important and precious business asset. Across industries and roles, we have all gained so much from data that it’s hard to imagine working without it anymore. But data, while valuable, isn’t like other corporate assets. When we talk about the values of business assets, we’re usually talking about a quantitative market value. You could liquidate your accounts, sell items around the office, and even put a price tag on intellectual property. Your data has no inherent value. When you’re talking about data, value isn’t in the data itself. 

So what do we mean when we talk about data value?  

Data value definition 

Let’s start with the definition of data itself. Data refers to a set of raw qualitative or quantitative variables. (While “datum” is the valid singular form of data in many fields, in computer sciences we typically refer to the singular and plural as “data.”) Data objects can take the form of measurements, statistics, or nearly any other attribute of people, places, or things. This raw data generally has a very low value.

Across the data life cycle, analytical processes make raw data useful by translating those attributes into information or intelligence with context and a business purpose. Decision-ready data and analytics inform business activities. When those activities save money or make money for an organization, they’ve finally unlocked that data’s business value. 

With that, we can define data value as the measurable financial impact of how your organization applies that data

Types of data value 

There are many ways that data can bring value to an organization. To measure financial impacts of data, look for specific ways the organization uses data to reduce costs, as well as ways that data drives revenue. Here are a few types of data-driven activities with measurable financial impact: 

  1. Gaining operational efficiency through transparency Shared, trusted data breaks down silos between different departments and provides real-time visibility into activity across the organization, with partners, or with vendors in your supply chain. 
    • Example: Manufacturers cut maintenance costs and minimize downtime using real-time sensor data and predictive analytics. By better anticipating equipment failure, they can plan timely repair or replacement.  
  2. Making better use of human resources with automationFrom email vacation responders to robot vacuums, these days we're all automating something to make our lives easier. Automated processes save people tedious steps at work, improving employee productivity. 
    • Example: Banks and financial institutions automate the preparation of data entered by customers, reserving employees’ valuable time for customer service tasks that require human expertise. 
  3. Reaching new audiences with segmentation and customizationGartner found that 63% of marketers struggle with personalized marketing. Data opens the door to customer segmentation and analyses that really help you understand who your customers are. 
    • Example: Field marketers integrate data from sources ranging from CRM to social media feeds. They use this to create personal, engaging experiences tailored to particular customer segments and markets. 
  4. Improving customer satisfaction with customer 360 As operations scale up and customer communications flood in, a data-driven 360-degree view of the customer helps bring back a human touch.
    • Example: Retailers integrate data from online and offline channels. Whenever customers visit in person, shop online, or call, they’ll get a cohesive customer experience that drives personal connection and loyalty.
  5. Innovating through augmented research and development Big data is a game-changer for R&D. Artificial intelligence is much better than we are at detecting the signal in the noise of vast amounts of data. Machine learning (ML) can make new discoveries using historical data, or deliver real-time insights.
    • Example: Consumer apps analyze user behavior to build product intelligence. That information inspires new features and informs go-to-market strategy for product updates and new products.  

Monetization of data 

In addition to using data to fuel revenue-driving activities from sales to product innovation, organizations can use data to generate value directly. Gartner describes data sharing as a key activity that’s increasingly necessary to succeed in the Big Data world. With the right approach, innovative companies have been able to turn their competitors into paying customers. They do this by packaging prepared data itself as a product or service.  

Monetizing data as a service (DaaS) is easier said than done. Delivering data to external customers raises the bar for data quality analysis and data availability. While any organization should already aim for infrastructure that provides internal customers with ready access to highly trusted data, making data externally available can reveal flaws. Issues that a company may have been willing to put up with for internal users could be dealbreakers for data monetization.

Before starting to share data externally, it’s also important to think carefully about the repositories where data is stored and managed. You may not want to give data customers access to a data warehouse where all your corporate data is stored. Instead, it would be wise to partition the monetized data, for example into a data mart. That way, the data for sale is safely isolated away from sensitive or proprietary data you don’t intend to share.  

How do I uncover data value? 

As we’ve discussed, you can derive value from data by using it to reduce costs, increase revenues, or generate income. In any case, the way to uncover your data’s net value is by calculating your return on investment. Capturing, moving, preparing, and storing data is not free. In order to determine your ROI on data, you’ll need to measure its costs and benefits.  

Investment in data is always an investment in the business. The impact of data investment, though, can vary wildly depending on where in the data life cycle you invest in data. Back in 1992, George Labovitz and Yu Sang Chang developed the 1-10-100 Rule for data costs: 

  • $1: cost to verify or standardize data at the point of entry 
  • $10: cost if you wait to cleanse the data until it’s in your system 
  • $100: cost of damage control if data is used without being cleansed 

Remember too that return on data can’t be measured by data volume, data speed, or even data quality alone. To be valuable, data has to be accessible to the people who need it, when they need it — a state we call data health.  

For example, pharmaceutical leader AstraZeneca achieved a stunning 40:1 ROI on their digital transformation with Talend. With a data catalog powered by Talend Data Fabric, 90 percent of the life science company’s data can now be ready for analysis within 3 minutes. This shaved a month off the time it takes AstraZeneca to run clinical trials, saving $1 billion per year. By making data more accessible across the company, they literally made their data more valuable to the business. For every $1 they spend on data, they now get $40 back. 

Do you have the metrics you need to assess data value at your organization? Would you like to improve your data ROI? Talend can help. Talend is on a mission to deliver the metrics necessary for organizations to measure data health. Customers like AstraZeneca are proving the value of this approach. Diagnosing problems that make data harder to use at your organization could be the first step to improving your data’s value. Contact us today to get started. 

Ready to get started with Talend?